The present invention relates to a grain sorting method and a device thereof which is used to continuously inspect and compare the appearance quality of the grains for sorting the same.
With respect to the grains such as rice and wheat, after harvested, the grains must be further processed before consumption. With rice exemplified, the rice grain must be first de-husked into brown rice. The brown rice then is processed into milled rice by removing its bran. Generally, the quality of rice is inspected with the brown rice as a standard. However. the existing rice inspecting method is applied by man. The objectivity of manual inspecting and the manual grading efficiency are major problems. Thus, it is necessary to develop a new method and machine to solve these problems. The existing rice inspecting machines only include color separation machine and single rice kernel sorting machine. These two machines will be described as follows:
1. The color separation machine is used in the end of a rice milling process for separating the abnormally colored rice kernel from the polished rice kernel. The milled rice kernels arranged as rows fall into in an inspecting room in which two sets of illuminating device and photoactive identifying device are installed. A fluorescent tube in the inspecting room is used to illuminate the rice kernels passing through. A color board is provided as a background for the rice kernels. A sensing section serves to detect the reflection light and refraction light of the rice kernel. Due to difference from the background color, the abnormally colored rice grain will be sensed by the photoelectric sensors. A controller will then activate an air nozzle for blowing high speed air to remove the abnormally colored rice kernel and only the polished rice kernel can pass through.
2. The single kernel sorting machine is used to identify five qualities of good (sound) kernel, immature kernel, damaged kernel, dead kernel and discolored kernel from each other. The machine is composed of a passing head, light splitting head and fissuring head. After the rice kernels are illuminated by the light, according to the gray level data obtained by the three detectors, the passing ratio, light splitting ratio and fissuring extent of the rice kernels are measured. Then a microcomputer generally processes and compares the detected value and grade standard value. Thereafter, the five classifications of good (sound) kernel, immature kernel, damaged kernel, dead kernel and discolored kernel are identified from each other. Finally, according to the identification result and the mixing ratio, the rice kernels are sorted into four grades.
However, the color separating machine employs the photoelectric sensor to compare the color difference between the milled rice kernels so as to remove the abnormally colored ones. This measure is not suitable for inspection of quality of rice or other grains. On the other hand, the single kernel sorting machine also employs the photoelectric measure to detect the color of the rice kernels but cannot detect the shape difference among the rice kernels. Therefore, the brown rice can be only substantially divided into five classifications and it is impossible to accurately sort the rice kernels. Therefore, the single kernel sorting machine is only applicable to preliminary sorting for brown rice. Moreover, such machine can only identify one single rice kernel at one time.